NUDGE - NUtritional, Digital Games in EnableHanna Schäfer

hanna.schaefer@tum.deDavid A. Plecherplecher@in.tum.de

Sophie L. Holzmannsophie.holzmann@tum.de

Georg Grohgrohg@in.tum.de

Gudrun Klinkerklinker@in.tum.de

Christina Holzapfelchristina.holzapfel@tum.de

Hans Haunerhans.hauner@tum.de

Technical University ofMunichGermany

Figure 1. Teaser figure of the NUDGE platform. Left: Central avatar element and the nutritional feedback screen. Middle: Motivational nudges andsocial cooking events. Right: Preview of the serious game prototype "Fit, Food, Fun".

ABSTRACTThe digital games platform NUDGE (NUtritional, DigitalGames in Enable) is one of the digital nutritional commu-nication systems within the nutrition research cluster enable.NUDGE aims at changing the nutritional behavior of adoles-cents using different levels of persuasion. The central platformelement is a personalized avatar representing the user’s nutri-tion and game status as playful feedback. Additionally, realworld interactions are integrated using tailored nudges, socialconnections, and real life challenges. Finally, different seriousgames, e.g. the "Fit, Food, Fun" prototype, are used to increasethe user’s knowledge and capabilities regarding nutrition. TheNUDGE platform is going to be evaluated in a pilot study inspring 2018.

Positive Gaming: Workshop on Gamification and Games for Wellbeing A CHI PLAY’17 Workshop, October 15, 2017, The Netherlands c© 2017 Copyright held by theauthors/owners.

Author KeywordsNutrition; Health; Adolescents; Gamification; Serious Games;Persuasion; Personalization; enable-Cluster;

INTRODUCTIONOverweight and obesity, especially in the younger population,are major health concerns worldwide (WHO, [34]). Accord-ing to a European statistic more than 60% of children that areaffected by overweight before puberty will maintain weight inyoung adulthood (WHO, [33]).The interdisciplinary enable-cluster is one of four clustersabout nutrition research funded by the German Federal Min-istry of Education and Research (BMBF) [14]. In one of itscore projects, enable investigates digital communication strate-gies to improve nutrition behavior. As a result, the NUDGEsystem is designed as an educative and preventive tool for ado-lescents to impart nutritional knowledge in a playful and moti-vational setting. To reach this goal, it promotes health-relatedoutcomes using serious games [1] and different persuasiveelements [27].Besides the imparting of nutritional knowledge, NUDGE triesto accompany its users by creating positive feedback loopsusing both automated and real world input to ensure the usersfeel comfortable with their behavioral changes.

PREVIOUS WORK ON HEALTH GAMESThere have been several studies on different types of gamesfor health and fitness. [3] reviewed 25 video games and foundthat they were split into different outcomes such as knowledgeincrease and changes in attitude and behavior. Since NUDGEwants to encompass a holistic approach the following six wellcited serious games were reviewed regarding their persuasivestrategies and health-related goals.One of the first digital nutritional games, ’Packy & Marlon’,was designed for children and adolescents with diabetes [8].The playing children had to manage the blood glucose levelsof two elephants with diabetes by choosing adequate food andinsulin quantities. A six-month randomized controlled trial(n=59; age=8-16 years) showed that after playing the user’sknowledge about diabetes increased, while the amount of ur-gent hospital visits decreased.The game ’HungryRedPlanet’ [2] (age=10-15 years) useda similar approach for healthy adolescents by utilizing theUSDA (U.S. Department of Agriculture) Pyramid Guide fora healthy and balanced nutrition. Similar to the game ’Packy& Marlon’, the goal of this game is to manage adequate foodquantities to win the game and stop earths starvation.The two serious games ’Escape from Diab’ (2010) [32](age=10-12 years) and ’PLAY, MATE!’ (2010) [4] (n=180;age=9-12 years) focused on persuasion and behavior changein addition to the knowledge transfer. The study of [4] showedthat games and persuasive elements could increase the physi-cal activity without disturbing the enjoyment of players.More health games are starting to integrate different persuasiveconcepts to address different target groups. For example, theNational Mindless Eating Challenge (NMEC) [19] (n=2053;age=39,8 years in average) integrated comparisons, customiza-tion, rewards, personalization and suggestions. Finally, thegame ’Move2Play’ [5] (n=12; age=12-13 years) provides asetup for physical activity similar to the one NUDGE is im-plementing for nutrition. They integrated monitoring, goal-setting, a social component, achievements, rewards, and anavatar component. Their small study showed, that the chil-dren were able to understand all different components of thesetup and enjoyed the customizing of the avatar and the socialcomponents.

CONCEPTS OF GAMES FOR HEALTHIn addition to the games review, different theoretical ap-proaches influenced the design of the NUDGE platform. Theconcept of serious gaming was applied to derive mini-gamesfor nutritional knowledge and skill transfer. The theory ofpersuasive gaming was used to develop the interaction of theplatform such as the avatar and the notification system. Theconcept of positive gaming was applied to furthermore im-prove the user’s wellbeing during his behavior change.

Serious GamingGamification as a preventive tool for promoting health-relatedbehaviors often lacks in efficacy. It mainly transfers gameelements such as points or leaderboards to a non-gaming con-tent [12]. In contrast to gamification, serious games offerthe possibility for an active exploration of serious intellectualproblems [1]. For example, a gamification approach would

be to give the user achievements, when he fills his nutritionaldiary. A serious game however would incorporate the nutri-tional knowledge and offer activities, that will transfer thatknowledge actively to the user. At the same time the gamingcharacter is stimulating the user’s intrinsic motivation to playand to learn [36]. The primary purpose of a serious game isto transfer knowledge in a playful gaming context [11]. TheNUDGE platform uses the concepts of serious gaming andmotivational stimulation in different nutritional mini-games.

Persuasive GamingPersuasive strategies are application components that are de-signed to influence the user’s behavior or attitude towards apredefined goal [16]. [27] defined a framework with 28 differ-ent persuasive strategies. [35] reviewed the implementation ofthose different strategies in 44 papers from different domains.For the health domain, they came up with the following tenmost used strategies: feedback, self-monitoring, suggestion,social role, simulation, tailoring, tunneling, reminders, reduc-tion, and reward.There are drawbacks to some of these strategies regarding theirapplicability to all user groups. Some strategies only work forspecific selection of personality types. [28] and [20] describehow using generic persuasive strategies might even reverse themotivational effect for specific personality types. However,strategies such as competition, comparison, self-monitoringand suggestion seem to be effective on most personality types[30]. The NUDGE platform uses the concepts of feedback,self-monitoring, social role, simulation, tailoring, reminders,rewards and personalization.

Positive GamingThe concept of positive computing was defined by Calvoand Peters (2014) [9] and includes three different design ap-proaches. The preventative design methodology excludes anynegative influence of the systems towards the user, while theactive design methodology tries to integrate elements, thatspecifically use positive influence on the user. Finally, thededicated design approach focuses on applications that havewellbeing as their primary goal. To implement those designstrategies [9] also suggest different factors such as positiveemotions, motivation, self-awareness, and empathy that canbe utilized when building positive applications. The NUDGEplatform uses the concepts of preventative and active designusing different intra- and interpersonal factors such as socialcomponents and a quantified-self avatar.

PRELIMINARY WORKThe target group of the NUDGE platform are adolescentsaged from 14 to 17 years. In preparation of the nutritionalgame design a survey was conducted. The aim was to de-scribe needs, wishes and motives of German adolescents re-garding nutritional and digital games, in order to design atarget group-specific serious game fostering an improved nu-trition. Literature [13][15] indicates that nutrition informationin adolescents is deficient and needs to be promoted. In 2010,the nutrition report of the German Nutrition Society (DGE)revealed that only about 50% of adolescents are adequately in-formed about nutrition [13]. The questionnaire was developed

through an interdisciplinary team of nutritionists and computerscientists. Socio-demographic and anthropometric data wereobtained as well. The 43-item questionnaire has been carriedout in the form of hard-copy and could be completed in 15to 20 minutes. In total, 339 adolescents aged from 13 to 18years (46% female), participated in the survey, which has beenconducted in six general secondary schools of the city andadministrative district of Rosenheim between June 2016 andJune 2017. The survey revealed that the nutritional knowledgein German adolescents is deficient. At same time, most ado-lescents play digital games and desire nutrition informationin digital games. Results are in line with findings of anothersurvey (2016) [15]. According to this survey two third ofthe 12 to 19 years old participants play digital games dailyor several times a week. Taken together, a serious nutritionalgame seems to be a potential educative and preventive tool toimpart nutritional knowledge in order to promote a healthylifestyle in adolescents.

SYSTEM DESIGNThe NUDGE platform is built as a system connecting multipleintervention bricks. The core of the platform is a digital avatar.It represents the health state of the user based on a digitalnutrition diary. The body composition of the avatar reflectsweight gain and loss. In addition to the physically adaption, theavatar also expresses moods such as hungry, lonely or excited.These moods should motivate the user to perform specificplatform actions such as filling his nutrition diary, interactingwith groups, or playing serious games. The avatar can enterdifferent serious mini-games such as the game prototype "Fit,Food, Fun", which focuses on macro-nutrients and energy con-tent of food. Other platform games address physical activityor micro-nutrient effects. Additionally, the user is motivatedby a social network and real life challenges, such as groupcooking events. He receives nudges about progress events andsocial comparisons. The following components try to includethe before mentioned persuasive or positive strategies.

Avatar as Central Platform ElementWithin the avatar different persuasion concepts are combined.First, the avatar can be adapted to the user’s appearance usingdifferent hair, clothing and color schemes. The designs werespecifically chosen to create an emotional connection. Thisdesign should increase the caretaking motivation between theusers and the avatar using a schema of childlike characteristics[17]. At the same time the avatar visualizes the nutrition statusof the user based on a dietary record. If the user eats healthyfor more than three days, the avatars bodyweight will decreaseand new clothing designs will be available. To prevent weightgain from causing negative psychological effects, the bodysizes are limited to 5 stages and were carefully selected in away that positively depicts even the heavier avatar versions.Besides this visual feedback, the user can also experimentwith nutritional effects using a fast forward scenario. Thisscenario shows the nutrient contents and their implications,in the given future dietary situation [25]. Both approachesutilize the self-quantification and simulation strategies of thepersuasive system design. Regarding positive computing theavatar should externalize the user’s emotions concerning his

healthy/unhealthy behavior and thus ease negative feedbackgiven by the system, while still giving motivation throughpositive feedback. Finally, the avatar serves as a bus betweenthe different serious games. For example, the avatar spendsenergy by playing the serious mini-games to limit the screentime for the participating adolescents. While the user’s nu-trition based on the dietary record affects the avatars’ bodycomposition, the user’s real life physical activity measured byGoogle Fit API impacts the avatar’s energy level [24]. Thisphysical activity is used to increase the avatar’s energy andthus represents another real-life connection. Further avatarcharacteristics, such as the current nutrient level will be inte-grated in future serious games e.g. by decreasing visibility orslowing movement. Besides the simulation state of the avatarregarding real life values such as nutrition and activity, theavatar also shows the platform status. For example, the avatarwill show hunger, if the user has not updated the diary, andit will be excited, if the user has not played any games in acertain time frame. This status feedback serves as a nudgebased on the user’s motivation through self-competition [10].

Social and Motivational ElementsSince the NUDGE platform tries to encompass real-life behav-ior change, the virtual avatar and serious games componentsmight not be sufficient to guide the users in their transition.Additional persuasive and positive elements that were incorpo-rated are personality dependent nudges, social support groupsand real-life challenges.According to [29] 6 of the 21 reviewed persuasive games fornutrition include a social support strategy. In the NUDGE plat-form, social support is given by groups of 5-10 people, thatshare challenges, a leaderboard and access to social games,such as the ISS space game. In case of missing social contactswith similar problems, the system will recommend groupsbased on age, location, gender and BMI. The group systemincluding the group size, group management and leaderboardwas evaluated in a small focus group with five participants anda usability test with five further participants was conductedas well [23]. Both evaluation methods revealed a positivefeedback on the general approaches, while the game designand usability were still criticized.It has been shown that tailoring nudges and personalizingachievements can strongly increase the motivational effect[28]. Our personality dependent nudges are based on sevengamer types identified by the BrainHex model [26] and the BigFive personality types [18]. Those types are used to select thetype of nudges users receive. For example, someone who is aseeker in the BrainHex model, would receive praise nudges ondiary entries, physical activity or weight loss. In addition toevent nudges, there are achievement nudges such as playingall games, or regularly filling the dietary diary.The real-life group challenges are integrated in the form ofsocial cooking events. This kind of challenge solves multipleissues of real-life challenges. First, it validates the user’s actionby the social feedback. Secondly, it offers social motivationfor the user to win the challenge. Finally, it opens a platformfor exchange about the received nutritional knowledge andother platform elements. These real-life group events wereevaluated in a small user study [23] with nine participants,

that conducted six different group cooking events within twoweeks. The study showed, that young people are interestedin such events. The most effective motivational elements arethe leaderboard list, showing the best cooking event/host ofeach group, the motto (e.g. Blue Titanic Dinner), giving eachcooking event a creative aspect, and the ingredient challenges,telling the host to e.g. involve salmon in his dinner.

"Fit, Food, Fun" as a Serious GameThe NUDGE platform combines different serious games suchas an ISS space game or a role-play exergame. This chapterexemplary describes one of the serious game prototypes called"Fit, Food, Fun" [37]. This game concentrates on improvingthe nutritional knowledge and on strengthening skills in usingthat knowledge to change dietary behavior. The knowledgetransfer is embedded in a journey through Europe. In everycountry the user has to succeed in three minigames:The Card Game offers the player two different cards with asort of food each. The challenge is to find the card with thehighest or lowest value in a certain category like carbs, fats,proteins and calories.In the Estimation Game the user has to guess the amount ofsugar or fat (displayed as cubes of sugar and drops of fat) thatis contained in a specific type of food.In the Backpack Game the player must fill a backpack withfood which should supply enough energy for a certain activitylike swimming or running. Without enough energy, the playerwon’t make it to the finish line. Choosing too much energyslows down the player resulting in a lower score.The first two games impart knowledge about the nutrientsof food whereas in the third game the user learns about theequivalence between physical activity and these nutrients.In addition to the link of nutritional knowledge and gamingstrategies, this serious game also integrates different edu-cational concepts [6]. For example, the transition betweencountries as well as the transition within the mini-games ofone country were designed along the AVIVA [31] model forteaching. Additionally, the game elements have been checkedagainst the ARCS model [21] and redefined to cover at leastone of the four motivational design elements. For example,the different country levels were restructured into a travelroute through Europe and complemented with informationabout each region to increase the curiosity of players.

CONCLUSION AND FUTURE WORKThe NUDGE platform is research project in progress. It triesto encompass a holistic set of persuasive and positive mea-sures into a serious game platform for health. While the majordesign and implementation steps are concluded, each com-ponent still needs to be improved through small focus groupinterviews. Also, the results of the preliminary survey will beincluded until the final evaluation. The final NUDGE platformis going to be evaluated in a pilot study in spring 2018. Dif-ferent outcomes like nutrition knowledge and motivation willbe assessed using appropriate questionnaires. To compare thecharacteristics of participants, the Big Five types [18] and theBrainHex gamer types [26] will be derived. Additionally, phe-notypic data (gender, age, body weight, body height, hip and

waist circumferences) will be obtained. The study outcomewill be assessed with different validated questionnaires suchas the User Experience Questionnaire [22] and the System Us-ability Scale [7]. Additionally, a domain specific survey willbe used to evaluate the knowledge and skill gain in nutrition.Finally, motivational questionnaires will be used pre and poststudy to show any changes in the users attitude. We expectedthe pilot study to show how knowledge gain and behaviorchange can be increased through persuasive and positive gameelements. Furthermore, we want to show that the users feelsafe and motivated during their behavioral transition due tothe social and motivational platform framework.

ACKNOWLEDGEMENTSThe preparation of this paper was supported by the enablecluster and is catalogued by the enable steering committee asenable 008 (http://enable-cluster.de). This work was fundedby a grant of the German Ministry for Education and Research(BMBF) FK 01EA1409H.

REFERENCES1. Clark C Abt. 1987. Serious games. University press of

America.

2. D Michael Anderson. 2002. Hungry Red Planet:Children’s multimedia nutrition simulation software. InThe 130th Annual Meeting of APHA.

3. Tom Baranowski, Richard Buday, Debbe I Thompson,and Janice Baranowski. 2008. Playing for real: videogames and stories for health-related behavior change.American journal of preventive medicine (2008), 74–82.

4. Shlomo Berkovsky, Mac Coombe, Jill Freyne, and DipakBhandari. 2010. Isn’t it great?: you can PLAY, MATE!.In Proceedings of the 15th international conference onIntelligent user interfaces.

5. Pavol Bielik, Michal Tomlein, Peter Krátky, ŠtefanMitrík, Michal Barla, and Mária Bieliková. 2012.Move2Play: An Innovative Approach to EncouragingPeople to Be More Physically Active. In Proceedings ofthe 2Nd ACM SIGHIT International Health InformaticsSymposium.

6. Alexandra Brandl. 2017. Design and Evaluation ofEducational Elements in a Serious Game on Nutrition.Technical University of Munich (2017).

7. John Brooke and others. 1996. SUS-A quick and dirtyusability scale. Usability evaluation in industry (1996).

8. Stephanie J Brown, Debra A Lieberman, BA Gemeny,Yong Chan Fan, DM Wilson, and DJ Pasta. 1997.Educational video game for juvenile diabetes: results of acontrolled trial. Medical Informatics (1997).

9. Rafael A Calvo and Dorian Peters. 2014. Positivecomputing: technology for wellbeing and humanpotential. MIT Press.

10. Zhi-Hong Chen, Tzu-Chao Chien, and Tak-Wai Chan.2011. Using Self-competition to Enhance Students’Learning. In International Conference on Technologies

for E-Learning and Digital Entertainment. Springer,234–235.

11. Michael David and Chen Sande. 2006. Serious Games:Games That Educate, Train, and Inform. Coursetechnology, Cengage learning (2006).

12. Sebastian Deterding, Dan Dixon, Rilla Khaled, andLennart Nacke. 2011. From game design elements togamefulness: defining gamification. In Proceedings of the15th international academic MindTrek conference:Envisioning future media environments.

13. DGE. 2000. Nutrition Report 2000. Henrich Druck +Medien GmbH.

14. enable cluster. 2017. Enable-Cluster.http://enable-cluster.de/. Accessed: 2017-08-11.

15. Sabine Feierabend, Theresa Plankenhorn, and ThomasRathgeb. 2016. JIM Studie 2016, Jugend, Information,(Multi-) Media Basisstudie zum Medienumgang 12- bis19-Jähriger in Deutschland. MedienpädagogischerForschungsverbund Südwest (mpfs).

16. B. J. Fogg. 2002. Persuasive Technology: UsingComputers to Change What We Think and Do. Ubiquity(2002).

17. Melanie L Glocker, Daniel D Langleben, Kosha Ruparel,James W Loughead, Ruben C Gur, and Norbert Sachser.2009. Baby schema in infant faces induces cutenessperception and motivation for caretaking in adults.Ethology (2009).

18. Oliver P John, Eileen M Donahue, and Robert L Kentle.1991. The big five inventory-versions 4a and 54. Berkeley,CA: University of California, Berkeley, Institute ofPersonality and Social Research.

19. Kirsikka Kaipainen, Collin R Payne, and Brian Wansink.2012. Mindless eating challenge: retention, weightoutcomes, and barriers for changes in a public web-basedhealthy eating and weight loss program. Journal ofmedical Internet research (2012).

20. Maurits Kaptein, Boris De Ruyter, Panos Markopoulos,and Emile Aarts. 2012. Adaptive persuasive systems: astudy of tailored persuasive text messages to reducesnacking. ACM Transactions on Interactive IntelligentSystems (TiiS) (2012).

21. John M Keller. 2009. Motivational design for learningand performance: The ARCS model approach.

22. Bettina Laugwitz, Theo Held, and Martin Schrepp. 2008.Construction and evaluation of a user experiencequestionnaire. In Symposium of the Austrian HCI andUsability Engineering Group.

23. Maximilian Alexander Mayer. 2017. Development andEvaluation of Food-Recommendation-System for Groups.Technical University of Munich (2017).

24. Sanjay M Mishra. 2015. Google Fit API. WearableAndroidTM: Android Wear & Google Fit AppDevelopment (2015).

25. Carsten Mück. 2017. Motivator Avatar: SupportingHealthy Nutrition. Technical University of Munich(2017).

26. Lennart E Nacke, Chris Bateman, and Regan L Mandryk.2014. BrainHex: A neurobiological gamer typologysurvey. Entertainment computing (2014).

27. Harri Oinas-Kukkonen and Marja Harjumaa. 2009.Persuasive systems design: Key issues, process model,and system features. Communications of the Associationfor Information Systems (2009).

28. Rita Orji, Regan L Mandryk, Julita Vassileva, andKathrin M Gerling. 2013. Tailoring persuasive healthgames to gamer type. In Proceedings of the SIGCHIConference on Human Factors in Computing Systems.

29. Rita Orji and Karyn Moffatt. 2016. Persuasive technologyfor health and wellness: State-of-the-art and emergingtrends. Health informatics journal (2016).

30. Rita Orji, Julita Vassileva, and Regan L Mandryk. 2014.Modeling the efficacy of persuasive strategies fordifferent gamer types in serious games for health. UserModeling and User-Adapted Interaction (2014).

31. Christoph Städeli. 2013. Kompetenzorientiertunterrichten-das AVIVA-Modell: fünf Phasen gutenUnterrichts.

32. Debbe Thompson, Tom Baranowski, Richard Buday,Janice Baranowski, Victoria Thompson, Russell Jago, andMelissa Juliano Griffith. 2010. Serious video games forhealth: How behavioral science guided the developmentof a serious video game. Simulation & gaming (2010).

33. WHO. 2008. Health topics - Noncommunicable diseases -Obesity - Data and statistics.http://www.euro.who.int/en/health-topics/

noncommunicable-diseases/obesity/data-and-statistics.Accessed: 2017-08-11.

34. WHO. 2014. Global status report on noncommunicablediseases 2014. World Health (2014).

35. Isaac Wiafe and Keiichi Nakata. 2012. BibliographicAnalysis of Persuasive Systems: Techniques; Methodsand Domains of Application. In Persuasive Technology:Design for Health and Safety; The 7th InternationalConference on Persuasive Technology; Sweden; June 6-8;Adjunct Proceedings.

36. Wee Ling Wong, Cuihua Shen, Luciano Nocera, EduardoCarriazo, Fei Tang, Shiyamvar Bugga, HarishkumarNarayanan, Hua Wang, and Ute Ritterfeld. 2007. Seriousvideo game effectiveness. In Proceedings of theinternational conference on Advances in computerentertainment technology.

37. Johannes Ziegltrum. 2017. Development of a SeriousGame to support a healthy diet. Technical University ofMunich (2017).